Great Article! Thanks for sharing the FFS framework. Just want to learn more about it, whether the foundational tech has to be digital or AI/ML related. Like the Moderna case, the NGS sequencing tech could be a foundational tech to identify the antigens further coded by the mRNA, or the NGS should be categorized as "scalability infrastructure"?
Hello, Jie! Thank you for your comment, good question. Well, the way I think about it is that foundational technology is "domain-agnostic." For instance, deep learning is foundational because it can be applied to pretty much anything, from financial modeling and weather forecasting to drug discovery. In contrast, functional technologies are domain-specific. For example, NGS is in my mind domain-specific, because it is only related to DNA/RNA and while it seems like a really broad thing in biology, but apart from that it is not applicable.
Regarding scalability tech/infrastructure, what I mean by that is the ability to scale laboratory process to a sort of semi-industrial or industrial scale. For example, there might be a great and sensitive technology for detecting COVID. But can you put it on a scanner in the airport to be able to screen millions of passengers per week? Can you make this lab method into a small device, and reusable for many interactions? If not, the technology is unlikely to become a strong business, no matter how sensitive and good. So, scalability tech is really about going from lab to high throughput/industrial realm. Not every lab breakthrough can be industrialized or even run at scale.
To give you an example of what I describe as scalability infrastructure is, for instance, machine learning-compatible chips/GPUs where you can run large language models at scale. Cloud infrastructure can be regarded as scalability technology, I suppose.
So, for example, a startup that runs mass spectrometry studies on patient samples to identify a newly discovered type of metabolites (functional technology) and analyzes them using deep learning to creative a disease model (foundational technology) absolutely requires the access to cloud infrastructure and computational resources to actually grow as business. If the startup does not have sufficient access or resources to scale the modeling project, I as an investor, would be worried, even though the metabolomics part is a breakthrough scientifically speaking.
Having said that, I welcome you to brainstorm and suggest improvements. FFS framework is evolving and it is just a way of thinking about something, it is not a quantitative framework, but a qualitative one.
Thank you so much for the detailed explanation! ：） In that case, I guess that each segment of the FFS model could be further divided into a sub-level FFS frame, just like NGS could be a foundational tech for all the DNA/RNA-related applications. it's great to learn the model.
Great Article! Thanks for sharing the FFS framework. Just want to learn more about it, whether the foundational tech has to be digital or AI/ML related. Like the Moderna case, the NGS sequencing tech could be a foundational tech to identify the antigens further coded by the mRNA, or the NGS should be categorized as "scalability infrastructure"?
Hello, Jie! Thank you for your comment, good question. Well, the way I think about it is that foundational technology is "domain-agnostic." For instance, deep learning is foundational because it can be applied to pretty much anything, from financial modeling and weather forecasting to drug discovery. In contrast, functional technologies are domain-specific. For example, NGS is in my mind domain-specific, because it is only related to DNA/RNA and while it seems like a really broad thing in biology, but apart from that it is not applicable.
Regarding scalability tech/infrastructure, what I mean by that is the ability to scale laboratory process to a sort of semi-industrial or industrial scale. For example, there might be a great and sensitive technology for detecting COVID. But can you put it on a scanner in the airport to be able to screen millions of passengers per week? Can you make this lab method into a small device, and reusable for many interactions? If not, the technology is unlikely to become a strong business, no matter how sensitive and good. So, scalability tech is really about going from lab to high throughput/industrial realm. Not every lab breakthrough can be industrialized or even run at scale.
To give you an example of what I describe as scalability infrastructure is, for instance, machine learning-compatible chips/GPUs where you can run large language models at scale. Cloud infrastructure can be regarded as scalability technology, I suppose.
So, for example, a startup that runs mass spectrometry studies on patient samples to identify a newly discovered type of metabolites (functional technology) and analyzes them using deep learning to creative a disease model (foundational technology) absolutely requires the access to cloud infrastructure and computational resources to actually grow as business. If the startup does not have sufficient access or resources to scale the modeling project, I as an investor, would be worried, even though the metabolomics part is a breakthrough scientifically speaking.
Having said that, I welcome you to brainstorm and suggest improvements. FFS framework is evolving and it is just a way of thinking about something, it is not a quantitative framework, but a qualitative one.
Thank you so much for the detailed explanation! ：） In that case, I guess that each segment of the FFS model could be further divided into a sub-level FFS frame, just like NGS could be a foundational tech for all the DNA/RNA-related applications. it's great to learn the model.
Makes perfect sense, thank you.